Research On Moving Target Tracking Method Based On Multi-camera Cooperation

Most of the application scenarios of traditional video surveillance technology use a single camera to detect and track moving targets.When using this method to track the target,the monitoring field of view is small,the level of intelligence of the monitoring equipment is low,and it takes more time.long.When the monitoring field of view is large,the field of view overlaps the field of view,and it is difficult to achieve real-time continuous tracking of the same moving target.Through the feature point matching and fusion of the field of view images in the multi-view field of view,the correlation between the overlapping field of view images is established,and then the target handover of the field of view is completed,and the moving target tracking accuracy of the multi-view video surveillance is improved.Therefore,this paper extracts and matches the feature points of the overlapped field of view,establishes a multi-view collaborative video surveillance overlapped field of view image field boundary line,studies the visibility of the moving target in the field of view,and conducts the target handover process.The identification solves the problem of multi-view coordinated overlapping view range moving target detection and tracking,and achieves real-time detection and tracking of overlapping view range moving targets.The specific research content of this article is as follows:(1)Completed the feature point matching of the video image of the multi-view collaborative overlapping view area.By improving the traditional overlapping view area video image feature point matching method,the image feature points in the overlapping view area are accurately extracted,and the overlapping image is effectively divided.,Increase the correlation between the overlapping view area images,realize the accurate matching of the feature points of the overlapping view area range,and provide a new idea and method for the multi-view collaborative overlapping view area video image feature point matching.(2)The boundary line of the visual field between the overlapping visual fields in the video image is established,and the handover of moving objects is realized.The traditional multi-view collaborative video surveillance image cannot effectively realize the handover of moving targets in the overlapping field of view scene,which easily leads to the loss of moving target tracking,and cannot realize the real-time continuous tracking of the moving target in the overlapping field of view.In this paper,the proposed feature point extraction method is used to obtain the same background image feature points in the overlapped field of view,and the relationship between the image feature points is used to establish the field of view boundary line between the overlapped fields of view,and realize the multi-view cooperative overlapped field of view field of view.The effective distinction.(3)Realize the real-time tracking of the same moving target,improve the traditional moving target detection and tracking method,eliminate the ghosting problem generated in the moving target tracking process,and effectively distinguish when there is an interference target that moves at the same time as the tracking target,Can realize real-time continuous tracking of moving targets,and provide a new idea and method for multi-view moving target detection and tracking.By setting up a comparison test,this paper conducts a comparison test from two aspects of the moving target tracking effect and the moving target tracking accuracy.In terms of target tracking effect,the overlapping view multi-view tracking algorithm can successfully achieve target handover and continuous tracking when tracking moving targets in three different scenarios.In terms of target tracking accuracy,the accuracy of the algorithm in this paper is 76%,87%,and 88%,respectively.The accuracy of the target tracking algorithm is higher than that of the improved algorithm in the literature.The method proposed in this paper can not only solve the real-time tracking of the same moving target,but also solve the problem that the moving target is lost due to the short tracking distance of the moving target in the multi-view cooperative overlapping field of view,and effectively improves the accuracy of target tracking.Multi-view moving target tracking provides a new solution.